© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    5. Llama3.1-8B-IT
    7. Resid Post - 131k
    9. 11-RESID-POST-AA
    11. 129908
    Prev
    Next
    INDEX
    Explanations

    mathematical solutions

    np_max-act · gemini-2.0-flash

    This neuron detects LaTeX‐style mathematical tokens—variable names (like X, T, ϕ, λ), function notations, and equation symbols—typically used when writing out differential‐equation solutions.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    andyrdt/saes-llama-3.1-8b-instruct/resid_post_layer_11/trainer_1
    Dataset (Dashboard)
    Various
    Features
    131,072
    Data Type
    float32
    Hook Name
    blocks.11.hook_resid_post
    Architecture
    standard
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    reau
    -0.06
     sails
    -0.06
     pady
    -0.06
     Sith
    -0.06
     ubiqu
    -0.06
    \Log
    -0.06
    -offset
    -0.06
    agrant
    -0.06
     Сов
    -0.06
    amar
    -0.05
    POSITIVE LOGITS
     Executes
    0.08
     initialValues
    0.08
    .models
    0.07
     José
    0.07
     Hearing
    0.07
    Rightarrow
    0.06
     closer
    0.06
     đậu
    0.06
     میتوان
    0.06
    _timeline
    0.06
    Activations Density 0.012%

    No Known Activations